A-FPN算法及其在遥感图像船舶检测中的应用

光学遥感图像船舶检测主要面临两个挑战:光学遥感图像背景复杂,船舶检测易受海浪、云雾及陆地建筑等多方面干扰;遥感图像分辨率低,船舶目标小,对于其分类与定位带来很大困难;针对上述问题,在FPN的基础上,提出一种融入显著性特征的卷积神经网络模型A-FPN (Attention-Based Feature Pyramid Networks)。首先,利用卷积提取图像特征金字塔;然后,利用顶层金字塔逐级构建显著特征层,抑制背景信息,通过金字塔顶层的细粒度特征提高浅层特征的表达能力,构建自上而下的多级显著特征映射结构;最后利用Softmax分类器进行多层级船舶检测。A-FPN模型利用显著性机制引导不同感受下的特征进行融合,提高了模型的分辨能力,对遥感图像处理领域具有重要应用价值。实验阶段,利用公开的遥感目标检测数据集NWPU VHR-10中的船舶样本进行测试,准确率为92.8%,表明A-FPN模型适用于遥感图像船舶检测。     

Discovery of Four New Clusters in the Cygnus Cloud

We report the discovery of four new open clusters(named QC 1,QC 2,QC 3 and QC 4)in the direction of Cygnus Cloud and select their members based on five astrometric parameters(l,b,ω,μα*,μδ)of Gaia DR2.We also derive their astrophysical parameters for each new cluster.Structure parameters are generated by fitting the radial density distribution with a King’s profile.Using solar metallicity,we performed isochrone-fitting on their purified color-magnitude diagrams(CMDs)to derive the age of the clusters.The known cluster NGC 7062 in an adjacent area is chosen to verify our identification process.The estimated distance,reddening and age of NGC 7062 are in good agreement with the literature.     

顾及地类转换差异的城市空间扩展元胞自动机模型及应用研究

元胞自动机模型已经成为城市空间扩展模拟研究的重要方法之一,并得到广泛应用。然而,现有的城市扩展元胞自动机模型仍存在不足。由于元胞状态设置较为简单,从而使模型转换规则中对不同用地类型向城市用地转换的差异与强度考虑不够。基于此本文在元胞自动机模型的框架下,设计了多元结构的元胞状态及转换规则,提出了顾及地类转换差异与强度的城市扩展元胞自动机模型。在计算非城市用地向城市用地转换的转换概率时,该模型考虑了3个方面的概率：① 地形地貌、经济发展等城市发展的驱动因素对城市用地扩展的影响概率,该概率采用logistics方法进行计算;② 邻域元胞的用地类型对中心元胞转换概率的影响,该概率采用扩展摩尔型方法进行计算;③ 不同类型的非城市用地（本研究中包括耕地、林地和裸地3种类型）向城市用地转换的强度,该概率由模拟基期土地利用数据与目标年份土地利用数据的叠加,得出不同类型的非城市用地在此时间段内向城市用地转换的规模,进而确定不同类型的非城市用地向城市用地转换的强度。最后,将以上3种概率的乘积作为元胞转换的概率。通过转换概率与转换阈值的对比判断中心元胞是否在下一个阶段转换为城市用地。经过迭代计算,不断增加城市用地元胞的数量。当模拟城市用地的结果与目标年份的城市用地规模差值在一定的范围内时停止模拟,得出最终结果。模型构建完成后,本文以长株潭城市群核心区为例进行了模拟实验。以2001年该地区的土地利用数据为基期数据,模拟2010年该地区的城市用地规模和空间分布。研究结果表明,根据本文提出的模型模拟的城市扩展结果与真实数据相比具有较高的一致性。模拟结果正确率达到68.66%,比基于传统logistics回归的元胞自动机模型的模拟精度提高了4.25%,Kappa系数为0.675。该模型较好地模拟了长株潭城市群核心区城市扩展,在城市空间扩展模拟中具有较好的适应性与有效性。     

Classification of stellar spectra with SVM based on within-class scatter and between-class scatter

Support Vector Machine (SVM) is a popular data mining technique, and it has been widely applied in astronomical tasks, especially in stellar spectra classification. Since SVM doesn’t take the data distribution into consideration, and therefore, its classification efficiencies can’t be greatly improved. Meanwhile, SVM ignores the internal information of the training dataset, such as the within-class structure and between-class structure. In view of this, we propose a new classification algorithm-SVM based on Within-Class Scatter and Between-Class Scatter (WBS-SVM) in this paper. WBS-SVM tries to find an optimal hyperplane to separate two classes. The difference is that it incorporates minimum within-class scatter and maximum between-class scatter in Linear Discriminant Analysis (LDA) into SVM. These two scatters represent the distributions of the training dataset, and the optimization of WBS-SVM ensures the samples in the same class are as close as possible and the samples in different classes are as far as possible. Experiments on the K-, F-, G-type stellar spectra from Sloan Digital Sky Survey (SDSS), Data Release 8 show that our proposed WBS-SVM can greatly improve the classification accuracies.     

民间信仰载体的地域格局及其影响因素——以兰州市榆中县祠庙文化为例

以祠庙祭祀为主体且祠庙保存良好的民间信仰是甘肃陇中地区民间文化的重要组成。甘肃陇中地区的民间信仰具有很强的趋同性，本文以陇中地区的榆中县为例，以各村的祠庙为民间文化载体，通过核密度分析、Logistic回归等方法探讨榆中县民间文化载体的空间分布及其影响因素对于探讨陇中地区人地关系，保护和弘扬民俗文化有着重要意义。研究结果表明：（1）根据已有研究将榆中县民间信仰归纳为山神信仰、水神信仰、天地信仰、女性信仰、英雄崇拜和祖先崇拜，祖先崇拜的祠庙数量占有重要地位；（2）榆中县祠庙主要分布在西北黄河南岸、中部陇海铁路沿线及南部风景名胜集聚区，村级层面空间差异不显著、乡镇空间差异相对较大。通过核密度分析，不同类别民间信仰空间分布热点各有不同；（3）榆中县民间信仰空间分布受到区位条件限制，民间信仰的祠庙多位于海拔高度较低、人口密度较大、交通可达性较好的地区，坡度和到水源的距离成为山神信仰和水神信仰祠庙空间分布显著的影响因子，榆中县祠庙呈现出山神信仰类祠庙“依山”，水神信仰类祠庙“傍水”的空间特点。     

苏北浅滩沉积物中的大型绿藻微观繁殖体的垂直分布和物种多样性

本研究通过在2017年秋冬季组织实施的2个绿潮种源调查航次,研究苏北浅滩紫菜养殖区沉积物中的大型绿藻微观繁殖体的垂直分布和物种多样性。对沉积物中的微观繁殖体的定量实验结果表明:表层泥样中绿藻微观繁殖体的丰度最高,随着深度的增加迅速降低。在9月份表层沉积物中丰度为9±2ind./g,而在11月份表层丰度达到98±25ind./g;与9月份相比, 11月份浅滩沉积物中绿藻繁殖体数量呈现出显著升高的趋势。通过对51个绿藻样品的分子鉴定发现:沉积物中存在石莼属(Ulva)、尾孢藻属(Urospora)、盘苔属(Blidingia)等大型绿藻的微观繁殖体;数量占优是石莼属繁殖体,主要包括曲浒苔(Ulva flexuosa)、浒苔(Ulva prolifera)、扁浒苔(Ulva compressa)、缘管浒苔(Ulva linza);其中曲浒苔占据的比例最高,在9和11月份分别达到80.94%和73.33%。在2个调查航次中,都从沉积物中发现了绿潮浒苔的微观繁殖体,这些微观繁殖体是绿潮浒苔在苏北浅滩种源维持的重要方式和关键阶段,也构成了黄海浒苔绿潮连年暴发的"种子库"。     

辽宁省海洋文化景观概述

文章介绍海洋文化景观的属性、定义和构成;将辽宁省海洋文化景观资源分为海洋物质文化景观和非物质文化景观2个大类,其中海洋物质文化景观包括海洋自然文化景观、海洋遗迹文化景观、海洋聚落文化景观、海洋宗教文化景观、海洋旅游文化景观和海洋设施景观,海洋非物质文化景观包括海洋历史文化景观、海洋民俗文化景观和海洋艺术文化景观,分别概述具体类别资源的代表内容;提出可采用模糊综合评价法评价辽宁省海洋文化景观,研究海洋文化景观对于海洋文化建设和提高海洋意识等具有重要意义。     

赤潮治理方法综述

文章综述目前赤潮治理的主要方法,其中物理法重点介绍机械搅动法、超声波法、吸附法和气浮法,化学法重点介绍无机药剂法、有机药剂法和胶体絮凝沉淀法,生物法主要介绍引入赤潮藻类生物天敌、微生物技术和化感技术,分析各自优势和不足,采用微生物技术和化感技术具有重要意义和发展前景。     

Study of the flood control scheduling scheme for the Three Gorges Reservoir in a catastrophic flood

The Three Gorges Project is the world's largest water conservancy project. According to the design standards for the 1,000‐year flood, flood diversion areas in the Jingjiang reach of the Yangtze River must be utilized to ensure the safety of the Jingjiang area and the city of Wuhan. However, once these areas are used, the economic and life loss in these areas may be very great. Therefore, it is vital to reduce this loss by developing a scheme that reduces the use of the flood diversion areas through flood regulation by the Three Gorges Reservoir (TGR), under the premise of ensuring the safety of the Three Gorges Dam. For a 1,000‐year flood on the basis of a highly destructive flood in 1954, this paper evaluates scheduling schemes in which flood diversion areas are or are not used. The schemes are simulated based on 2.5‐m resolution reservoir topography and an optimized model of dynamic capacity flood regulation. The simulation results show the following. (a) In accord with the normal flood‐control regulation discharge, the maximum water level above the dam should be not more than 175 m, which ensures the safety of the dam and reservoir area. However, it is necessary to utilize the flood diversion areas within the Jingjiang area, and flood discharge can reach 2.81 billion m³. (b) In the case of relying on the TGR to impound floodwaters independently rather than using the flood diversion areas, the maximum water level above the dam reaches 177.35 m, which is less than the flood check level of 180.4 m to ensure the safety of the Three Gorges Dam. The average increase of the TGR water level in the Chongqing area is not more than 0.11 m, which indicates no significant effect on the upstream reservoir area. Comparing the various scheduling schemes, when the flood diversion areas are not used, it is believed that the TGR can execute safe flood control for a 1,000‐year flood, thereby greatly reducing flood damage.